# Goal-oriented Object Importance Estimation in On-road Driving Videos

## TL;DR

This paper introduces Object Importance Estimation (OIE) for on-road driving videos, combining visual cues and driving goals to identify critical objects influencing vehicle control decisions.

## Contribution

It presents a novel framework integrating visual dynamics and driving goals for OIE, along with a real-world dataset and annotations for evaluation.

## Key findings

- Goal-oriented method outperforms baselines
- Significant improvement in turn scenarios
- Object importance enhances driving control prediction

## Abstract

We formulate a new problem as Object Importance Estimation (OIE) in on-road driving videos, where the road users are considered as important objects if they have influence on the control decision of the ego-vehicle's driver. The importance of a road user depends on both its visual dynamics, e.g., appearance, motion and location, in the driving scene and the driving goal, \emph{e.g}., the planned path, of the ego vehicle. We propose a novel framework that incorporates both visual model and goal representation to conduct OIE. To evaluate our framework, we collect an on-road driving dataset at traffic intersections in the real world and conduct human-labeled annotation of the important objects. Experimental results show that our goal-oriented method outperforms baselines and has much more improvement on the left-turn and right-turn scenarios. Furthermore, we explore the possibility of using object importance for driving control prediction and demonstrate that binary brake prediction can be improved with the information of object importance.